Compositional Vector Space Models for Knowledge Base Completion
This work addresses the problem of enhancing knowledge base inference for AI applications by enabling more flexible and generalizable reasoning, though it is incremental in advancing existing compositional methods.
The paper tackles knowledge base completion by developing a compositional vector space model that uses a recursive neural network to reason about conjunctions of multi-hop relations, enabling generalization to unseen paths and zero-shot learning for new relation types. It shows improvements of 11% over a traditional classifier and 7% over a method with pre-trained embeddings on a new dataset of over 52M relational triples.
Knowledge base (KB) completion adds new facts to a KB by making inferences from existing facts, for example by inferring with high likelihood nationality(X,Y) from bornIn(X,Y). Most previous methods infer simple one-hop relational synonyms like this, or use as evidence a multi-hop relational path treated as an atomic feature, like bornIn(X,Z) -> containedIn(Z,Y). This paper presents an approach that reasons about conjunctions of multi-hop relations non-atomically, composing the implications of a path using a recursive neural network (RNN) that takes as inputs vector embeddings of the binary relation in the path. Not only does this allow us to generalize to paths unseen at training time, but also, with a single high-capacity RNN, to predict new relation types not seen when the compositional model was trained (zero-shot learning). We assemble a new dataset of over 52M relational triples, and show that our method improves over a traditional classifier by 11%, and a method leveraging pre-trained embeddings by 7%.